Business Problem

No-Churn Telecom is an establish Telecom operation in Europe with more than a decade in Business. Due to new players in the Market, telecom industry has become very competitive and retaining customers becoming challenge. In spite of No-Churn initiative for reducing tariffs and promoting more offers, the churn rate (percentage of customers migrating to competitors) is well above 10%. **No-Churn wants to explore possibility of Machine Learning to help with following use cases to retain competitive edge in the industry.

Goal

Help No-Churn with their use cases with ML 1. Understanding the variables that influencing the customers to migrate. 2. Creating Churn risk scores that can be indicative to drive retention campaigns. 3. Introduce new predicting variable “CHURN-FLAG” with values YES(1) or NO(0) so that email campaigns with lucrative offers can be targets to Churn YES customers. 4. Exporting the trained model with prediction capability for CHURN-FLAG Highlights the flag (with input variables documents) that can be integration with internal application help to identify possible CHURN-FLAG YES customers and provide more attention in customer touch point areas, including customercare support, request fulfilment, auto categorizing tickets as high priority for quick resolutions any questions they may have etc.,

Import Libraries

As there are no missing values in the data we are directly going for some exploratory data

Exploratory Data Analysis

Lets Visualize the data with different types of plots and also determine the relation ship between them

Assigining the Input and Output Variables

For Assigining inputs wefirst need to go through a scalar and also encode the categorical values

Builiding the Model

Importing the required library

Now here we are checking the relative importance of the input columns with respect to the output columns as shown below

Conclusion

we conclude the project by summarizing the goals reached as per the requrirement of the project 1. First goal is to understand the variables that influence the customers to migrate. --If we check the flow of the people that want to migrate the most influencing factors are Day charge and Customer service as you can see people mostly depend on the cost high is always a no and you can see that in the Day charge graph in the plot --For customer service calls as number of calls increase you can see that more people are leaving as they are not satisfied with the service provided. --Another influence factor is the International plan as it takes a different cost charge and call time than the regular calls as the data is given as such so it induces a different Charge and more of the international plan yes people are migrating 4.Created a predicted model with a accuracy score of 96 for predicting whether churn_Flag is yes or no